The excipients added to therapeutic protein formulations are more specific than biomanufacturers typically realize. As it turns out, these inactive ingredients have markedly different effects on solubility, and those differences are based on certain specific physicochemical factors.
In a recent paper, researchers from the University of Cambridge, Eindhoven University of Technology, and Merck identified solubility differences among commonly used antibodies that ranged from approximately 7-fold to 200-fold. The differences came down to molecular specificity and the complexity of the antibody-excipient interactions. By understanding these indicators and their effects, biomanufacturers can design mAb formulations using a more predictive, rational design approach.
A team of eight, including first author Zexiang Han, a PhD candidate in the Knowles lab; and senior authors Tuomas Knowles, PhD, professor; and Pietro Sormanni, PhD, all of the University of Cambridge, analyzed the effects of four common pharmaceutical excipients (histidine, sodium chloride, arginine, and sucrose) on the solubility of a panel of therapeutic immunoglobulin G (IgG) antibodies using a high-throughput droplet microfluidic platform. They determined that excipient-mediated solubility depends strongly upon the antibodies themselves, and that solubility “can be quantitatively linked to specific molecular features derived from sequence and structure.”
For example, when looking at the solubility of 1 mg mL−1 cetuximab in the presence of each of the four excipients, they reported that as excipient concentration increased, greater percentages of polyethylene glycol (PEG) crowder were needed to cause mAb precipitation. That was true for each of the excipients.
The scatter plots in the paper showed the relative concentrations of each excipient needed to meaningfully shift the solubility boundary. The histidine plot, for example, stops at about 35 mM, while the plot for sucrose stops at about 500 mM. For cetuximab, a relatively low concentration of histidine shifted the boundary, while a significantly higher concentration of sucrose was needed to cause a similar shift. The contrast among the various excipients, given the same antibody and same concentrations of PEG, underscores that the shifts were caused through molecularly specific interactions between the antibody and the excipient rather than by PEG’s excluded volume effect (i.e., crowding).
More specifically, they noted:
- “Histidine stabilization is fundamentally polar, correlating with antibody dipole moments.
- “Sodium chloride operates via nonspecific electrostatic screening to suppress self-association in low-pI variants [those in which the pI—the isoelectric point—is below that of the pH and have no charge].
- “Arginine efficiency reflects a competitive tradeoff between electrostatic screening and chaotropic effect.”
The takeaway, they reported, is that “formulation additives do not exert universal actions.” Instead, common excipients behave differently with different molecules. This work, therefore, provides “a mechanistic, physics-based framework that can support the rational design and optimization of antibody formulations,” the team pointed out.
